Chemokine receptors mediate the migration of leucocytes during inflammation. The cytoplasmic protein FROUNT binds to chemokine receptors CCR2 [chemokine (C-C motif) receptor 2] and CCR5, and amplifies chemotactic signals in leucocytes. Although the interaction between FROUNT and chemokine receptors is important for accurate chemotaxis, the interaction mechanism has not been elucidated. In the present study we identified a 16-amino-acid sequence responsible for high-affinity binding of FROUNT at the membrane-proximal C-terminal intracellular region of CCR2 (CCR2 Pro-C) by yeast two-hybrid analysis. Synthesized peptides corresponding to the CCR2 Pro-C sequence directly interacted with FROUNT in vitro. CCR2 Pro-C was predicted to form an amphipathic helix structure. Residues on the hydrophobic side are completely conserved among FROUNT-binding receptors, suggesting that the hydrophobic side is the responsible element for FROUNT binding. The L316T mutation to the hydrophobic side of the predicted helix decreased the affinity for FROUNT. Co-immunoprecipitation assays revealed that the CCR2 L316T mutation diminished the interaction between FROUNT and full-length CCR2 in cells. Furthermore, this mutation impaired the ability of the receptor to mediate chemotaxis. These findings provide the first description of the functional binding element in helix 8 of CCR2 for the cytosolic regulator FROUNT that mediates chemotactic signalling.

INTRODUCTION

The chemokine receptor family comprises 19 GPCRs (G-protein-coupled receptors). The C-terminal intracellular domain of chemoattractant receptors is indispensable for chemotaxis [1,2] and the subsequent receptor endocytosis [3,4]. In particular, the membrane-proximal region of chemokine receptors seems to play an essential role in chemotaxis [510].

The cytoplasmic protein FNT [FROUNT; also known as NUP85 (nuclear pore complex protein Nup85)] directly interacts with the C-terminal intracellular domain of activated CCR2 [chemokine (C-C motif) receptor 2] and amplifies the chemokine-elicited PI3K (phosphoinositide 3-kinase)–Rac–lamellipodium protrusion cascade, promoting subsequent chemotaxis. If the interaction between FNT and CCR2 is blocked, the chemotactic response is prevented [11], suggesting that the interaction between FNT and CCR2 plays a crucial role in CCR2-mediated chemotaxis.

We reported previously that FNT is a common regulator of CCR2 and another chemokine receptor CCR5 [12]. Although the functions of CCR2 and CCR5 differ owing to different chemokine usage, these receptors mediate the migration and infiltration of monocyte/macrophages and are involved in the pathogenesis and progression of cancer and inflammatory diseases, such as arteriosclerosis, multiple sclerosis and graft versus host disease [1316]. Thus the interaction between CCR2/CCR5 and FNT is a promising drug target for inflammatory diseases; however, the molecular mechanism underlying the binding of FNT to CCR2/CCR5 has not been elucidated.

To investigate the molecular mechanism of the interaction between CCR2 and FNT, in the present study we have identified a short sequence in the membrane-proximal C-terminal region of CCR2 (CCR2 Pro-C) responsible for efficient binding to FNT. This sequence is predicted to form a helical structure, and we have identified residues (Leu316 and His323) crucial for FNT binding on the hydrophobic side of the predicted helix. Interestingly, the residues on the hydrophobic side of CCR2 are identical to those of the other chemokine receptor CCR5 to which FNT binds. Thus in the present study we define the hydrophobic side of a 16-amino-acid putative helix of CCR2/CCR5 Pro-C as an element for high-affinity binding to FNT. Furthermore, co-immunoprecipitation assays revealed that a low-affinity L316T mutation in this binding element impairs the interaction between full-length CCR2 and FNT in cultured cells. This low-affinity mutation correlated with the impaired chemotaxis mediated by CCR2.

EXPERIMENTAL

Proteins and peptides

Full-length human FNT was expressed in Escherichia coli cells as a soluble protein, as described by Esaki et al. [17]. The C-terminal region of FNT (FNT-C) was fused to a GST tag and GST–FNT-C was expressed in E. coli BL21 cells (Merck Millipore) as a soluble protein. The cells were lysed, and the GST–FNT-C protein was purified using a glutathione–Sepharose 4B column (GE Healthcare). For preparation of unlabelled FNT-C, GST was excised from FNT-C by thrombin protease in the column. FNT-C was further purified using a Q Sepharose Fast Flow column, followed by Superdex 200 gel filtration chromatography (GE Healthcare). Peptides corresponding to the FNT-binding region of CCR2 and CCR5 or CXCR4 [chemokine (C-X-C motif) receptor 4; EKFRRYLSVFFRKHIT, EKFRNYLLVFFQKHIA or AKFKTSAQHALTSVSR respectively], and other mutants (L316T, EKFRRYTSVFFRKHIT and H323Y, EKFRRYLSVFFRKYIT; underlined residue indicates the mutation) were custom synthesized (Hayashi Kasei). In some experiments, peptides biotinylated at the N-terminal side were used.

Yeast two-hybrid system

Deletion mutants were generated from cDNAs encoding the C-terminal domain of CCR2 and subcloned into pAS2-1 GAL4-BD (Clontech). These clones and the pACT GAL4-AD vector (Clontech) containing a cDNA encoding amino acids 500–656 of human FNT were co-transfected into yeast Y190 cells using a lithium acetate protocol. Double-positive clones were selected by growth in selection medium lacking tryptophan (T) and leucine (L) (TL medium). Interactions between FNT and each deletion mutant of CCR2 in yeast cells were tested for histidine (H) auxotrophy (TLH medium) and β-galactosidase activity using a filter lift assay. A semi-quantitative liquid β-galactosidase assay using ONPG (o-nitrophenyl β-D-galactopyranoside) as a substrate was performed according to the protocol described for the MATCHMAKER System-2 (Clontech).

SPR analysis

The interactions between FNT and the CCR2 Pro-C peptide or various mutants of this peptide were analysed by SPR (surface plasmon resonance) using Biacore X and Biacore2000 instruments (GE Healthcare). FNT protein was immobilized on the CM5 sensor chip, and various concentrations of the CCR2 Pro-C peptide in HBS-EP buffer (GE Healthcare) were applied to the sensor chip at a flow rate of 20 μl per min and then washed out with HBS-EP buffer. For the experiment determining the binding ability of CCR5 Pro-C to FNT, CCR5 Pro-C peptide with two lysine residues added to the C-terminus was used. For the other experiments, the Pro-C peptide was immobilized on the CM5 sensor chip and FNT protein was applied to the sensor chip at the same flow rate. RUs (resonance units) were measured during the binding and washing periods, and the binding kinetics were analysed using BIAevaluation software version 3.0. Kinetic constants were obtained from the association and dissociation curves generated from different concentrations of analyte by using the 1:1 Langmuir binding model (A+BAB) available in the software. This model was found to fit the data very well when GST–FNT (Figure 2B), but not the CCR2 Pro-C peptide, was used as the analyte (Figure 2A). Use of a more complex binding model did not give a better fit of the data. All binding curves were corrected for background by subtraction of the reference flow cells.

HTRF-based binding assay

The HTRF (homogeneous time-resolved fluorescence) assay was performed in white 384-well low-volume microplates (Corning Coaster). To achieve the composition of the final reaction solution, 5 nM GST-fused FNT-C, 250 nM biotinylated Pro-C peptide, 2.6 ng of anti-GST antibody labelled with Europium Cryptate and 12.5 ng of high-grade XL665-conjugated streptavidin were mixed in HTRF buffer [10 mM Hepes (pH 7.4), 0.2 M potassium fluoride, 10 mM NaCl, 0.1% Tween 20 and 0.5% BSA]. As a negative control, a biotinylated peptide corresponding to the CXCR4 sequence was used. HTRF signals were measured at room temperature (25°C) using an Envision instrument (PerkinElmer) at 620 and 665 nm Em (emission) wavelengths. The results are expressed as a Em665nm/Em620nm ratio or Delta F calculated by eqn (1):

 
formula
(1)

In the competition assay, unlabelled FNT-C or full-length FNT and unrelated recombinant protein as a control were added at the indicated concentrations into the HTRF reaction mixture containing 5 nM GST–FNT and 250 nM CCR2 Pro-C peptide.

Cluster analysis of GPCRs

We retrieved 1272 vertebrate GPCR sequences from the SEVENS database (http://sevens.cbrc.jp) [17a] and compared them with UniProt/Swiss-Prot. The regions of transmembrane helices (TMH1–TMH7) and loop regions of these sequences were determined on the basis of the available crystal structures of bovine rhodopsin, adrenergic receptors and the adenosine receptor, to which multiple sequence alignments of the selected sequences were assigned with avoidance of gaps in the transmembrane domains. The multiple sequence alignment was modified by manual curation. The detailed conditions of the sequence comparison and detailed principles of the multiple structural alignment are described previously [18]. In the present analysis, 16 residues corresponding to positions 310–325 of the CCR2 sequence are defined as the Pro-C region. We selected the corresponding regions of 179 human GPCRs, including 19 chemokine receptors {CCR1–CCR10, CCRL1 (CCR-like 1), CXCR1–CXCR6, CX3CR1 [chemokine (C-X3-C motif) receptor 1] and XCR1 [chemokine (C motif) receptor 1]}, on the basis of the alignment. These amino acid sequences were translated to feature vectors with the following physico-chemical elements: molecular mass [19], charge, pK value [20], hydrophobicity [21] and number of hydrogen bonds [22] for each amino acid. The vectors of the short Pro-C region sequences and the full-length sequences were then analysed by a clustering method (Ward method). This methodology is adequate to link sequences that have similar physico-chemical properties even though they have no sequence similarity.

Semi-random mutagenesis and selection

To generate a mutational library of the CCR2 Pro-C region, semi-random mutagenesis of the pAS2-1 CCR2 vector was performed by following the guidelines of Sawano and Miyawaki [23] with the exception of the oligonucleotides TR448 (35-mer), 5′-CATG-GAGGCCGAATTCNNGAAGTTCAGAAGGTATC-3′; TR449 (34-mer), 5′-GGAGGCCGAATTCGAGNNGTTCAGAAGGTA-TCTC-3; TR450 (34-mer), 5′-GGCCGAATTCGAGAAGNN-CAGAAGGTATCTCTCG-3; TR451 (34-mer), 5′-CGAATTCG-AGAAGTTCNNAAGGTATCTCTCGGTG-3′; TR452 (33-mer), 5′-GAGAAGTTCAGAAGGNNTCTCTCGGTGTTCTTC-3′; TR453 (33-mer), 5′-GAAGTTCAGAAGGTATNNCTCGGTG-TTCTTCCG-3′; TR454 (32-mer), 5′-GGTATCTCTCGGTGNN-CTTCCGAAAGCACATC-3′; TR455 (31-mer), 5′-GTATCTCT-CGGTGTTCNNCCGAAAGCACATC-3′; TR456 (32-mer), 5′-CTCGGTGTTCTTCCGANNGCACATCACCAAGCG-3′; and TR457 (35-mer), 5′-CGGTGTTCTTCCGAAAGNNCATCACC-AAGCG-3′ that were used to generate the semi-randomized mutants. In the above oligonucleotide sequences ‘N’ designates the inclusion of equimolar concentrations of all of the nucleotides during oligonucleotide synthesis. More than 100 mutants were sequenced and 78 different clones were picked up. FNT-binding affinity was tested by an ONPG assay in yeast cells. High- and low-affinity mutated CCR2 Pro-C peptides were synthesized and selected by both SPR analysis and HTRF assay.

Co-immunoprecipitation assay

Human FROUNT cDNA (corresponding to amino acids 23–656) fused to EGFP, and human CCR2 (pcMGS-CCR2) or low-affinity mutant human CCR2 (pcMGS-CCR2-L316T) cDNA were co-transfected into HEK (human embryonic kidney)-293 cells with the Lipofectamine® reagent (Invitrogen). Transfection efficiency was confirmed by GFP fluorescence and CCR2 staining by an anti-(human phycoerythrin-conjugated CCR2) antibody (R&D Laboratories) using a Gallios flow cytometer (Beckman Coulter). More than 90% of cells were positive for CCR2 expression. These cells were stimulated with CCL2 (Peprotech) for 5 min and then lysed in a detergent buffer containing 50 mmol/l Tris/HCl buffer (pH 7.6), 150 mmol/l NaCl, 1% Nonidet P40, 0.5% sodium deoxycholate and Protease Inhibitor cocktail (Nakarai). Samples were then centrifuged at 10000 g for 10 min at 4°C. The soluble fraction was incubated at 4°C with a mouse anti-FLAG (M2) antibody coupled to Dynabeads® M-270 epoxy (Invitrogen) for 30 min. Immunoprecipitates were eluted and subjected to SDS/PAGE (4–12% gel), transferred on to PVDF membranes and stained with anti-EGFP (Clontech) or anti-FLAG antibody (Invitrogen).

Chemotaxis assay

Human Jurkat cells were maintained in RPMI 1640 medium supplemented with 10% FBS. Human CCR2 or CCR2-L316T (a low-affinity mutant) and the EGFP vector, as a marker for gene expression, were transiently co-transfected into Jurkat cells using a Nucleofector device (Amaxa) and a Nucleofection kit V according to the manufacturer's instructions. At 16 h after nucleofection, living cells at a density of 106 cells/ml were placed into the upper chamber of a 96-well chemotaxis chamber (Neuroprobe) with a polycarbonate filter (5-μm pore size). After incubation at 37°C for 120 min, EGFP-positive cells that had migrated to the lower chamber containing medium or CCL2 (Peprotech) were counted under a fluorescence microscope.

RESULTS

A 16-amino-acid sequence in the membrane-proximal region of the C-terminal cytoplasmic tail of CCR2 is responsible for efficient binding to FNT

The cytoplasmic chemotaxis regulator FNT has been identified as a molecule that binds to the C-terminal region of CCR2 in a Y2H (yeast two-hybrid) system [11]. To pinpoint the region needed for FNT binding, we examined a series of C-terminal-truncation mutants of CCR2 for their ability to bind FNT in Y2H assays (Figure 1A). The results are summarized in Figure 1(B). Truncation of the distal C-terminal region (amino acids 329–360) of CCR2 maintained the interaction with FNT (clone CCR2-0 in Figure 1B and [11]). A further 5-amino-acid deletion at the C-terminal end (Figure 1B, clones CCR2-1–CCR2-5) did not reduce the ability to bind FNT. Interestingly, truncation of a further three amino acids from 309–328 at the C-terminal end (Figure 1B, clone CCR2-3) led to increased binding ability in the β-galactosidase assay, whereas a further deletion of three amino acids (Figure 1B, CCR2-6) led to decreased FNT binding. Regarding the N-terminal end, deletion of more than two amino acids (Figure 1B, CCR2-9 and CCR2-13) completely diminished the ability to bind FNT. A semi-quantitative ONPG assay revealed that the most potent binding ability was exhibited by clone CCR2-12 (Figure 1C). Thus a 16-amino-acid region corresponding to residues 310–325 of CCR2 (CCR2 Pro-C) was defined as the high-affinity region for FNT.

Identification of a short sequence with a high affinity for FNT at the membrane-proximal C-terminal region of CCR2

Figure 1
Identification of a short sequence with a high affinity for FNT at the membrane-proximal C-terminal region of CCR2

(A) To pinpoint the region in the C-terminal domain of CCR2 responsible for FNT binding, sequential truncation mutants of the C-terminal domain of CCR2 were generated and assessed for their ability to bind FNT by Y2H analysis. (B) The FNT-binding ability of each truncation mutant was assessed by expression of a reporter gene measured using auxotrophy (TLH) and a colorimetric β-galactoside (β-gal) assays. (C) The binding ability of various lengths of the truncation mutants (underlined in B) was quantified by an ONPG assay. A 16-amino-acid region corresponding to residues 310–325 of CCR2 (CCR2 Pro-C; membrane-proximal C-terminal intracellular region) was defined as the high-affinity region for FNT (boxed area in B). Nega, control vector.

Figure 1
Identification of a short sequence with a high affinity for FNT at the membrane-proximal C-terminal region of CCR2

(A) To pinpoint the region in the C-terminal domain of CCR2 responsible for FNT binding, sequential truncation mutants of the C-terminal domain of CCR2 were generated and assessed for their ability to bind FNT by Y2H analysis. (B) The FNT-binding ability of each truncation mutant was assessed by expression of a reporter gene measured using auxotrophy (TLH) and a colorimetric β-galactoside (β-gal) assays. (C) The binding ability of various lengths of the truncation mutants (underlined in B) was quantified by an ONPG assay. A 16-amino-acid region corresponding to residues 310–325 of CCR2 (CCR2 Pro-C; membrane-proximal C-terminal intracellular region) was defined as the high-affinity region for FNT (boxed area in B). Nega, control vector.

The Pro-C sequence of CCR2 alone directly interacts with FNT

To confirm whether the identified region directly binds to FNT in vitro, a 16-amino-acid peptide corresponding to the Pro-C high-affinity region for FNT was synthesized and assessed for binding to the FNT protein by SPR. The CCR2 Pro-C peptide showed a dose-dependent response towards the C-terminal CCR2-binding region of FNT tagged with GST (GST–FNT-C) immobilized on a sensorchip, whereas the control (CXCR4 Pro-C) did not (Figure 2A). A reverse assay was performed in which the CCR2 Pro-C peptide was immobilized along with the CXCR4 peptide as a reference. The GST-fused FNT protein (but not GST alone) gave a dose-dependent binding response to the CCR2 Pro-C peptide (Kd=5.3 μM) (Figure 2B).

Interaction between FNT and the 16-amino-acid CCR2 ProC peptide

Figure 2
Interaction between FNT and the 16-amino-acid CCR2 ProC peptide

The synthetic CCR2 peptide corresponding to the FNT-binding domain (16-amino-acid sequence of CCR2 Pro-C) determined in Figure 1 was assessed for binding to the FNT protein. (A) SPR sensorgrams of CCR2 Pro-C peptide binding to the immobilized GST-fused FNT C-terminal fragment (FNT-C) are shown. A peptide derived from the CXCR4 sequence corresponding to the location of CCR2 Pro-C was used as a control (CXCR4 Pro-C, grey line). (B) SPR sensorgrams of GST-fused FNT-C (black line) and GST protein (grey line) binding to immobilized CCR2 Pro-C peptide. (C) The ability of the CCR2 Pro-C peptide to bind to FNT was examined by a HTRF assay using an indirect ‘cassette format’, in which the interaction between GST-fused FNT-C and the biotinylated CCR2 Pro-C peptide was detected by an anti-tag donor/acceptor pair, namely cryptate-labelled anti-GST antibodies and a streptavidin-XL665 conjugate respectively. GST protein and the biotinylated CXCR4 Pro-C peptide were used as controls. Data at 30 min after mixing are shown. (D) Interaction between CCR2 Pro-C and FNT at various concentrations of CCR2 Pro-C peptide. Data at 5 min after mixing are shown. (E) HTRF competition assay using unlabelled fragment (FNT-C) or full-length (FNT-Full) FNT protein. The competitors were added to the HTRF reaction mixture between GST-fused FNT-C and CCR2 Pro-C as described in (C). The percentage inhibition of the FNT–CCR2 Pro-C interaction caused by the addition of each competitor was calculated relative to the signal obtained from a negative control. Results are means ± S.D.

Figure 2
Interaction between FNT and the 16-amino-acid CCR2 ProC peptide

The synthetic CCR2 peptide corresponding to the FNT-binding domain (16-amino-acid sequence of CCR2 Pro-C) determined in Figure 1 was assessed for binding to the FNT protein. (A) SPR sensorgrams of CCR2 Pro-C peptide binding to the immobilized GST-fused FNT C-terminal fragment (FNT-C) are shown. A peptide derived from the CXCR4 sequence corresponding to the location of CCR2 Pro-C was used as a control (CXCR4 Pro-C, grey line). (B) SPR sensorgrams of GST-fused FNT-C (black line) and GST protein (grey line) binding to immobilized CCR2 Pro-C peptide. (C) The ability of the CCR2 Pro-C peptide to bind to FNT was examined by a HTRF assay using an indirect ‘cassette format’, in which the interaction between GST-fused FNT-C and the biotinylated CCR2 Pro-C peptide was detected by an anti-tag donor/acceptor pair, namely cryptate-labelled anti-GST antibodies and a streptavidin-XL665 conjugate respectively. GST protein and the biotinylated CXCR4 Pro-C peptide were used as controls. Data at 30 min after mixing are shown. (D) Interaction between CCR2 Pro-C and FNT at various concentrations of CCR2 Pro-C peptide. Data at 5 min after mixing are shown. (E) HTRF competition assay using unlabelled fragment (FNT-C) or full-length (FNT-Full) FNT protein. The competitors were added to the HTRF reaction mixture between GST-fused FNT-C and CCR2 Pro-C as described in (C). The percentage inhibition of the FNT–CCR2 Pro-C interaction caused by the addition of each competitor was calculated relative to the signal obtained from a negative control. Results are means ± S.D.

We further tested the interaction between the CCR2 Pro-C and FNT in solution by using a HTRF assay. The interaction was detected by using an indirect ‘cassette format’, in which the interaction between GST–FNT-C and the biotinylated CCR2 Pro-C peptide was detected by an anti-tag donor/acceptor pair, namely cryptate-labelled anti-GST antibody and a streptavidin-XL665 conjugate respectively. The CCR2 Pro-C peptide biotinylated at the N-terminal end used in this assay showed FNT binding comparable with that of unlabelled peptides in the SPR analysis (results not shown). In this assay, when the CCR2 Pro-C peptide was mixed with GST–FNT-C protein, but not with GST alone, a FRET signal was observed (Figure 2C). By contrast, no signal was observed when CXCR4 Pro-C was mixed with GST–FNT-C (Figure 2C). The FRET signal was observed within a few minutes when the CCR2 Pro-C peptide was mixed with GST–FNT in a dose-dependent manner (Figure 2D). The specific interaction of FNT and CCR2 Pro-C was confirmed by a competition assay (Figure 2E). An unlabelled FNT fragment (FNT-C) or full-length FNT inhibited this interaction dose-dependently, whereas the control protein did not.

Clustering analysis of various human GPCRs by Pro-C region highlights conservation among FNT-binding receptors

Identification of the 16-amino-acid short sequence for FNT binding in CCR2 prompted us to look for conservation of this sequence among known GPCRs. We generated dendrograms based on feature vectors of the full-length amino acid sequence and the 16-amino-acid Pro-C region (see the Supplementary Online Data at http://www.biochemj.org/bj/457/bj4570313add.htm). The dendrogram generated from feature vectors of the full-length sequences gave the expected clustering of receptor families; thus the chemokine receptor family (19 receptors) was found in the same clade. In the dendrogram generated from the short feature vectors of the Pro-C region, the topology of the dendrogram changed markedly, with the chemokine receptor family interspersed among the clades. Notably, however, CCR2 and CCR5 (to which FNT binds) were located proximally, whereas CXCR4 (to which FNT does not bind) appeared on a distant branch. These results indicate that CCR2 and CCR5 are the most homologous in the Pro-C sequence among all 179 of the examined GPCRs.

The Pro-C region of CCR2 may form an amphipathic helix and residues on the hydrophobic side of the predicted helix are responsible for FNT binding

To identify the key residues in CCR2 Pro-C responsible for the interaction with FNT, CCR2 Pro-C mutants were semi-randomly generated in a Y2H vector (Figure 3A), and 78 clones were tested for their ability to bind FNT using an ONPG assay (Figure 3B). Subsequently, those peptides with an amino acid mutation affecting FNT binding in the Y2H assay were synthesized and their binding strength was verified by SPR. Among the mutant CCR2 Pro-C peptides tested, substitution of Leu316 with a threonine residue (CCR2 Pro-C L316T) resulted in a marked reduction in binding to the immobilized FNT protein (Figure 4A, left-hand panel), whereas substitution of His323 with a tyrosine residue (CCR2 Pro-C H323Y) markedly increased the binding response to FNT (Figure 4A, right-hand panel).

Generation of Pro-C mutants to identify the key residues for the interaction with FNT

Figure 3
Generation of Pro-C mutants to identify the key residues for the interaction with FNT

(A) Experimental procedure to assess the FNT-binding affinity of CCR2 Pro-C point mutants generated by random mutagenesis using PCR. (B) CCR2 Pro-C mutants were tested for their ability to bind FNT by a high-throughput ONPG assay in yeast cells. Mutant H323Y and mutant L316T are indicated by arrows.

Figure 3
Generation of Pro-C mutants to identify the key residues for the interaction with FNT

(A) Experimental procedure to assess the FNT-binding affinity of CCR2 Pro-C point mutants generated by random mutagenesis using PCR. (B) CCR2 Pro-C mutants were tested for their ability to bind FNT by a high-throughput ONPG assay in yeast cells. Mutant H323Y and mutant L316T are indicated by arrows.

Binding of the mutated CCR2 Pro-C to FNT and role of the amphipathic helical structure in CCR2 Pro-C on FNT binding

Figure 4
Binding of the mutated CCR2 Pro-C to FNT and role of the amphipathic helical structure in CCR2 Pro-C on FNT binding

(A) SPR sensorgrams of synthetic peptides of native and mutated CCR2 Pro-C at the indicated concentration binding to immobilized FNT protein. (B) The confidence rate (Conf) for a helical secondary structure (0=low and 9=high) of the C-terminal region of CCR2 was calculated by the PSIPRED protein structure prediction server. The shaded area indicates the frame identified as the FNT-binding region. (C) Helical wheel projections of Pro-C sequences with low FNT affinity (L316T, left-hand panel), high FNT affinity (H323Y, right-hand panel) and native sequence (middle panel) (http://rzlab.ucr.edu/scripts/wheel/wheel.cgi). The hydrophobic residues are grey-coloured diamonds, hydrophilic residues are white circles, potentially negatively charged residues are white triangles and potentially positively residues charged are white pentagons. The asterisk indicates the mutated amino acid. (D) Interaction signals (Delta F) of native or the mutated CCR2 Pro-C biotinylated peptide with FNT in HTRF (means ± S.D.). (E) Effect of native or mutated CCR2 Pro-C unlabelled peptide added as a competitor in the interaction of biotinylated native CCR2 Pro-C peptide with FNT in HTRF (means ± S.D.).

Figure 4
Binding of the mutated CCR2 Pro-C to FNT and role of the amphipathic helical structure in CCR2 Pro-C on FNT binding

(A) SPR sensorgrams of synthetic peptides of native and mutated CCR2 Pro-C at the indicated concentration binding to immobilized FNT protein. (B) The confidence rate (Conf) for a helical secondary structure (0=low and 9=high) of the C-terminal region of CCR2 was calculated by the PSIPRED protein structure prediction server. The shaded area indicates the frame identified as the FNT-binding region. (C) Helical wheel projections of Pro-C sequences with low FNT affinity (L316T, left-hand panel), high FNT affinity (H323Y, right-hand panel) and native sequence (middle panel) (http://rzlab.ucr.edu/scripts/wheel/wheel.cgi). The hydrophobic residues are grey-coloured diamonds, hydrophilic residues are white circles, potentially negatively charged residues are white triangles and potentially positively residues charged are white pentagons. The asterisk indicates the mutated amino acid. (D) Interaction signals (Delta F) of native or the mutated CCR2 Pro-C biotinylated peptide with FNT in HTRF (means ± S.D.). (E) Effect of native or mutated CCR2 Pro-C unlabelled peptide added as a competitor in the interaction of biotinylated native CCR2 Pro-C peptide with FNT in HTRF (means ± S.D.).

Most GPCRs have an amphipathic helical structure, located proximal to the seventh transmembrane domain. CCR2 is also predicted to have a helical structure in this region, as revealed by the PSIPRED protein structure prediction server (http://bioinf.cs.ucl.ac.uk/psipred/) [23a] (Figure 4B). The 16-amino-acid Pro-C region identified for FNT binding lies completely within this helical region. Helical wheel projection revealed that CCR2 Pro-C contains a grey-coloured hydrophobic cluster on one face (Figure 4C, middle panel), suggesting that CCR2 Pro-C forms an amphipathic α-helical structure. Interestingly, both of the residues implicated in FNT binding by Y2H and SPR analysis are located in the hydrophobic cluster of CCR2 Pro-C (Figure 4C).

An HTRF assay also revealed the increased binding of the H323Y mutant and the decreased binding of the L316T mutant to the FNT C-terminal region (Figure 4D). Consistent with its binding affinity for FNT, unlabelled high-affinity mutant H323Y peptide competed more efficiently than the wild-type sequence with the labelled wild-type CCR2 pro-C peptide for FNT binding, whereas the low-affinity mutant peptide L316T failed to compete (Figure 4E). Taken together, these results indicate that the hydrophobic side of the 16-amino-acid CCR2 Pro-C helix is responsible for binding to FNT.

Residues on the hydrophobic side of the helix-prone Pro-C region are completely conserved among two FNT-binding receptors

FNT also binds to another chemokine receptor, CCR5, to promote the directional migration of cells [12]. To gain further insight into the key residues for binding to FNT, we compared the properties of the Pro-C 16-amino-acid sequence of CCR2 with that of CCR5. The C-terminal regions of CCR2 and CCR5 possess high similarity. Regarding the Pro-C 16-amino-acid sequence of CCR2 identified as FNT binding in the Y2H assay, the corresponding sequence of CCR5 is identical except for four amino acids. Helical wheel analysis showed that CCR5 Pro-C also exhibits an amphipathic α-helical structure (Figure 5A, middle panel). Note that this region of CXCR4, which does not interact with FNT in either the Y2H or SPR analysis, does not contain the highly hydrophobic cluster seen in CCR2 (Figure 5A, bottom panel). The four different amino acids in CCR5 are all located on the side opposite to the hydrophobic cluster, whereas the residues comprising the hydrophobic side are identical to those of CCR2 Pro-C (Figure 5A, top panel). As expected from the above observation, the CCR5 Pro-C peptide bound efficiently to FNT in the SPR analysis (Figure 5B). We also tested the ability of the 16-amino-acid peptide of CCR5 to bind to FNT in an HTRF assay (Figure 5C). The biotinylated 16-amino-acid CCR5 peptide gave a response comparable with that of CCR2 when mixed with FNT. Furthermore, the competition assay shows that the unlabelled CCR5 Pro-C peptide, but not the CXCR4 Pro-C peptide, inhibited the interaction between the biotinylated CCR2 Pro-C peptide and FNT (Figure 5D). These results indicate that the common residues on the hydrophobic side of CCR2 and CCR5 Pro-C define the receptors’ ability to bind to FNT.

Common residues on the hydrophobic side of CCR2 Pro-C and CCR5 Pro-C and their role on FNT binding

Figure 5
Common residues on the hydrophobic side of CCR2 Pro-C and CCR5 Pro-C and their role on FNT binding

(A) Helical wheel projections of the CCR2 pro-C region (amino acids 310–325) and the corresponding region of CCR5 and CXCR4. The hydrophobic residues are grey-coloured diamonds, hydrophilic residues are white circles, potentially negatively charged residues are white triangles and potentially positively charged residues are white pentagons. The asterisk indicates the amino acid distinct from that of CCR2. (B) The 16-amino acid peptides of CCR5, but not CXCR4, bound to the recombinant FNT-C protein in SPR analysis. (C) Interaction signals derived from the interaction between the biotinylated CCR2 Pro-C peptide or the biotinylated CCR5 Pro-C peptide and the recombinant FNT-C protein in a HTRF assay. (D) Competition assay using unlabelled CCR5 Pro-C or CXCR4 Pro-C at a concentration of 1 μM against the interaction of biotinylated CCR2 Pro-C peptide with FNT in a HTRF assay (means ± S.D.). *P<0.05.

Figure 5
Common residues on the hydrophobic side of CCR2 Pro-C and CCR5 Pro-C and their role on FNT binding

(A) Helical wheel projections of the CCR2 pro-C region (amino acids 310–325) and the corresponding region of CCR5 and CXCR4. The hydrophobic residues are grey-coloured diamonds, hydrophilic residues are white circles, potentially negatively charged residues are white triangles and potentially positively charged residues are white pentagons. The asterisk indicates the amino acid distinct from that of CCR2. (B) The 16-amino acid peptides of CCR5, but not CXCR4, bound to the recombinant FNT-C protein in SPR analysis. (C) Interaction signals derived from the interaction between the biotinylated CCR2 Pro-C peptide or the biotinylated CCR5 Pro-C peptide and the recombinant FNT-C protein in a HTRF assay. (D) Competition assay using unlabelled CCR5 Pro-C or CXCR4 Pro-C at a concentration of 1 μM against the interaction of biotinylated CCR2 Pro-C peptide with FNT in a HTRF assay (means ± S.D.). *P<0.05.

Mutation in the hydrophobic side of the helix-prone Pro-C region impairs the ability of whole CCR2 to interact with FNT and mediate chemotaxis

To confirm the role of the element of CCR2 identified for FNT binding in the context of the whole receptor molecule, we compared the FNT binding ability of full-length wild-type CCR2 and full-length CCR2 with a mutation corresponding to the low-affinity mutant of the Pro-C peptide identified by SPR and HTRF (CCR2-L316T) in a co-immunoprecipitation analysis. cDNA encoding human CCR2 or CCR2-L316T tagged with FLAG, and cDNA encoding EGFP-fused FNT were co-transfected into HEK-293 cells, and these receptors were expressed on the cell surface in equal amounts (Figure 6A). Immunoprecipitation was performed using the anti-FLAG antibody (Figure 6B), and co-immunoprecipitation of the FNT protein with CCR2 was detected by the anti-EGFP antibody (Figure 6C). Less FNT was co-immunoprecipitated with the CCR2-L316T mutant (Figures 6C and 6D), even though immunoprecipitation of the mutant receptor was equal to that of the wild-type receptor (Figure 6B). These results demonstrate that the key residue for binding to FNT identified in vitro, Leu316 in CCR2 Pro-C, has a crucial role in the full-length CCR2–FNT interaction in cells. Next, we evaluated the effect of the L316T mutation on CCR2-mediated chemotaxis. Wild-type CCR2 or the CCR2-L316T expression vector used in Figures 6(A)–6(D) was transfected into human Jurkat T-cells, which express endogenous FNT, but not CCR2. CCR2-transfected cells, but not cells transfected with control vector, exhibited chemotactic activity toward CCL2 (results not shown). Although CCR2-L316T was expressed on the cell surface comparably with the wild-type CCR2, chemotactic activity was significantly diminished in cells transfected with CCR2-L316 (Figure 6E). These results demonstrate that the key residue for binding to FNT identified in vitro, Leu316, has a crucial role in the full-length CCR2–FNT interaction in cells and in the receptor's ability to mediate chemotaxis.

FNT-binding ability of full-length, wild-type or mutated CCR2 and chemotactic ability of cells expressing mutated CCR2

Figure 6
FNT-binding ability of full-length, wild-type or mutated CCR2 and chemotactic ability of cells expressing mutated CCR2

(A) Human CCR2 or CCR2-L316T tagged with FLAG and EGFP-tagged FNT were co-transfected into HEK-293 cells. GFP fluorescence derived from EGFP–FNT and surface expression of FLAG–CCR2 were measured by flow cytometry (solid line). The broken line indicates no transfected cells. Mean fluorescence intensities (MFI) are shown in each panel. (B and C) Immunoprecipitation (IP) of CCL2-stimulated transfected HEK-293 cell lysates was performed using an anti-FLAG antibody followed by a Western blot assay (WB). Immunoblots were probed with the anti-FLAG antibody to detect immunoprecipitation of wild-type CCR2 or CCR2-L316T (B), and with the anti-EGFP antibody to detect co-immunoprecipitation of EGFP–FNT (C). The immunoblot of each cell lysate is shown below. Molecular mass (MW) is given on the left-hand side in kDa. (D) Co-immunoprecipitation data from three independent experiments were quantified and normalized to the amount of each receptor immunoprecipitated and are shown as the fold change over the wild-type. (*P<0.05). (E) Chemotaxis of Jurkat cells transfected with wild-type CCR2 (white bar) or CCR2-L316T (black bar). Cells that had migrated from the upper chamber to the lower chamber containing medium or CCL2 were counted under a fluorescence microscope. Results are means ± S.D. (*P<0.05 compared with the wild-type CCR2).

Figure 6
FNT-binding ability of full-length, wild-type or mutated CCR2 and chemotactic ability of cells expressing mutated CCR2

(A) Human CCR2 or CCR2-L316T tagged with FLAG and EGFP-tagged FNT were co-transfected into HEK-293 cells. GFP fluorescence derived from EGFP–FNT and surface expression of FLAG–CCR2 were measured by flow cytometry (solid line). The broken line indicates no transfected cells. Mean fluorescence intensities (MFI) are shown in each panel. (B and C) Immunoprecipitation (IP) of CCL2-stimulated transfected HEK-293 cell lysates was performed using an anti-FLAG antibody followed by a Western blot assay (WB). Immunoblots were probed with the anti-FLAG antibody to detect immunoprecipitation of wild-type CCR2 or CCR2-L316T (B), and with the anti-EGFP antibody to detect co-immunoprecipitation of EGFP–FNT (C). The immunoblot of each cell lysate is shown below. Molecular mass (MW) is given on the left-hand side in kDa. (D) Co-immunoprecipitation data from three independent experiments were quantified and normalized to the amount of each receptor immunoprecipitated and are shown as the fold change over the wild-type. (*P<0.05). (E) Chemotaxis of Jurkat cells transfected with wild-type CCR2 (white bar) or CCR2-L316T (black bar). Cells that had migrated from the upper chamber to the lower chamber containing medium or CCL2 were counted under a fluorescence microscope. Results are means ± S.D. (*P<0.05 compared with the wild-type CCR2).

DISCUSSION

Intracellular chemotactic signals mediated via the chemokine receptors CCR2 and CCR5 are positively modulated by their interaction with the cytosolic regulator FNT. FNT amplifies the chemokine-elicited PI3K–Rac–lamellipodium protrusion cascade; however, the molecular mechanism underlying this interaction has not been elucidated. In the present study, we pinpointed the receptor region responsible for binding to FNT and identified a short high-affinity amino acid sequence in the membrane-proximal C-terminal region of CCR2 (CCR2 Pro-C; residues 310–325). We demonstrated that the hydrophobic side of the helix predicted for this sequence is the element responsible for FNT binding using a mutational study and a comparison between FNT-binding and non-binding receptors. Despite their high homology, the function of these receptors seems to be different owing to different chemokine usage. In the present study, we propose that these different chemokine receptors are regulated by a common regulator via a common binding element.

The Y2H assay of a series of deletion mutants revealed that residues 310–325 exhibited a more efficient binding than either a longer peptide or the entire cytoplasmic tail (Figure 1C). The present study clearly shows the potential of this 16-amino-acid region of CCR2/CCR5 to interact with FNT (Figures 2 and 5). The direct interaction between a 16-amino-acid synthesized peptide, corresponding to CCR2 Pro-C, and the FNT protein was confirmed by two different approaches, SPR analysis and HTRF assay. The affinity of the interaction between FNT and CCR2 was estimated at approximately 5 μM (Figure 2B). The specificity of the interaction was demonstrated by competition with either the unlabelled full-length or the C-terminal domain of FNT, and by the lack of competition with the control protein (Figure 2E).

Furthermore, we confirmed the role of the identified FNT-binding element of CCR2 in the context of the whole receptor molecule (Figure 6). The L316T mutation, which impaired the interaction of CCR2 pro-C with FNT in Y2H, SPR and HTRF analyses, also impaired the ability of the whole CCR2 molecule to interact with FNT in human cultured cells. Moreover, we revealed that the mutation in the FNT-binding element of CCR2 dramatically impaired the receptor's ability to mediate cellular chemotaxis.

Identification of a short sequence for FNT binding in CCR2 enabled us to propose a new grouping of GPCRs on the basis of the 16-amino-acid sequence at the membrane-proximal C-terminal region corresponding to the FNT-binding region of CCR2 (see the Supplementary Online Data). Chemokine receptors are interspersed in the dendrogram based on this sequence. Nevertheless, no other GPCRs share more homology in the Pro-C region than CCR2 and CCR5 despite the shortness of the region. Intriguingly, this clustering analysis suggests that other neighbouring pairs of receptors might share different Pro-C-binding molecules.

There are several reports on cytoplasmic molecules binding via the Pro-C region. For example, in the case of CXCR2, a chemokine receptor expressed on neutrophils, LASP-1 [LIM and SH3 (Src homology 3) domain protein-1] has been reported to bind to the C-terminal region of the receptor [24]. The residues responsible for LASP-1 binding are located in the region corresponding to the 16-amino-acid FNT-binding site of CCR2/CCR5, but are not identical to those responsible for FNT binding, indicating that the sequence of this region defines the characteristics for binding cytoplasmic regulatory molecules.

Chemokine receptors are GPCRs containing seven transmembrane domains. Most structurally characterized GPCRs have helical structures in the membrane-proximal C-terminal region (helix 8). Indeed, the presence of helix 8 has been reported in a crystal structure of CCR5 [25]. The FNT-binding region of CCR2/CCR5 identified in the present study is located in the so-called helix 8 region. Our studies demonstrated that the helical state of the helix 8 region is responsible for binding to the cytoplasmic protein FNT (Figure 4); furthermore, residues on the hydrophobic side of the helix 8 region are completely conserved among the two FNT-binding receptors CCR2 and CCR5 (Figure 5). Previous reports have demonstrated important roles for the helix 8 region in receptor activation [2628], although the molecular mechanisms remain unclear. The results of the present study provide new insight into the regulation mechanism mediated by helix 8, in which the hydrophobic side of helix 8 serves as a binding site for a cytoplasmic regulator. Based on our findings, together with previous findings, we propose a hypothesis that FNT, upon chemokine stimulation, accumulates to the plasma membrane to interact with the helix 8 region of chemokine receptors, promoting receptor clusterization and amplifying receptor signalling, leading to accurate formation of pseudopodia directed towards the chemokine concentration.

Chemokine receptors have been considered as potential drug targets because they have been reported to contribute to the onset and progression of cancer and inflammatory diseases [29]. The diversity among chemokines and chemokine receptors has been well documented over the past two decades since the discovery of chemokine prototypes [3032]; by contrast, the diversity of cytoplasmic-binding molecules is not well understood. Besides its role in monocytes/macrophages, FNT is required for the migration and recruitment of CCR2-expressing bone marrow-derived mesenchymal stem cells to the injured heart [33]. FNT has also been reported to mediate the transendothelial migration of prostate carcinoma through activation of the small G-protein Rac [34]. Although the physiological dynamics of the receptor–FNT interaction should be further studied, our findings provide insight into the mechanism underlying the interaction of chemokine receptors with the cytoplasmic regulator FNT, and the interaction systems described in the present study would be useful tools for drug development targeting the interaction between CCR2/CCR5 and FNT.

In conclusion, our findings emphasize the importance of a short element at the membrane-proximal C-terminal region of chemokine receptors, and perhaps other GPCRs, for binding to cytoplasmic regulatory molecules. Identification of other proteins that bind to this region of other GPCRs may lead to further understanding of receptor activation mechanisms.

We thank Kana Kokubo, Yuko Kobayashi, Ryu Takahashi, Taeko Ito, Mizuho Ishida, Masaru Ueda, Saeko Ohi, Yasushi Hiramatsu, Jutaro Kawaguchi, Yukimi Nakayama and Jumpei Shosaku for experimental help, and all the members of our laboratory for discussions and help.

Abbreviations

     
  • CCR

    chemokine (C-C motif) receptor

  •  
  • CXCR

    chemokine (C-X-C motif) receptor

  •  
  • Em

    emission

  •  
  • FNT

    FROUNT

  •  
  • GPCR

    G-protein-coupled receptor

  •  
  • HEK

    human embryonic kidney

  •  
  • HTRF

    homogeneous time-resolved fluorescence

  •  
  • LASP-1

    LIM and Src homology 3 domain protein-1

  •  
  • ONPG

    o-nitrophenyl β-D-galactopyranoside

  •  
  • PI3K

    phosphoinositide 3-kinase

  •  
  • SPR

    surface plasmon resonance

  •  
  • Y2H

    yeast two-hybrid

AUTHOR CONTRIBUTION

Etsuko Toda and Yuya Terashima conceived the study, performed the experiments, analysed the data and wrote the paper. Kaori Esaki, Sosuke Yoshinaga and Hiroaki Terasawa prepared the FNT protein and contributed to discussions. Yutaka Kofuku and Ichio Shimada assisted in the conduction of the SPR experiments using Biacore2000 and contributed with important advice to the paper. Minoru Sugihara and Makiko Suwa conducted the cluster analysis of GPCRs in the Supplementary Online Data and assisted in preparing the paper. Shiro Kanegasaki critically reviewed the paper prior to submission and Kouji Matsushima gave conceptual advice. Yuya Terashima directed the overall research. All authors read and approved the final paper.

FUNDING

This work was supported, in part, by the MEXT (Ministry of Education, Culture, Sports, Science and Technology) Japan via the TPRP (Targeted Proteins Research Programme) and the P-DIRECT (Project for Development of Innovative Research on Cancer Therapeutics) and a Grant-in-Aid for Young Scientists (B) [grant number 24790462 (to Y.T.)], and the JSPS (Japan Society for the Promotion of Science) [grant number 24-40099 (to E.T.)]

References

References
1
Richardson
R. M.
Ali
H.
Pridgen
B. C.
Haribabu
B.
Snyderman
R.
Multiple signaling pathways of human interleukin-8 receptor A. Independent regulation by phosphorylation
J. Biol. Chem.
1998
, vol. 
273
 (pg. 
10690
-
10695
)
2
Ko
J.
Jang
S. W.
Kim
Y. S.
Kim
I. S.
Sung
H. J.
Kim
H. H.
Park
J. Y.
Lee
Y. H.
Kim
J.
Na
D. S.
Human LZIP binds to CCR1 and differentially affects the chemotactic activities of CCR1-dependent chemokines
FASEB J.
2004
, vol. 
18
 (pg. 
890
-
892
)
3
Sambrano
G. R.
Coughlin
S. R.
The carboxyl tail of protease-activated receptor-1 is required for chemotaxis. Correlation of signal termination and directional migration
J. Biol. Chem.
1999
, vol. 
274
 (pg. 
20178
-
20184
)
4
Fan
G. H.
Yang
W.
Wang
X. J.
Qian
Q.
Richmond
A.
Identification of a motif in the carboxyl terminus of CXCR2 that is involved in adaptin 2 binding and receptor internalization
Biochemistry
2001
, vol. 
40
 (pg. 
791
-
800
)
5
Ben-Baruch
A.
Bengali
K. M.
Biragyn
A.
Johnston
J. J.
Wang
J. M.
Kim
J.
Chuntharapai
A.
Michiel
D. F.
Oppenheim
J. J.
Kelvin
D. J.
Interleukin-8 receptor β. The role of the carboxyl terminus in signal transduction
J. Biol. Chem.
1995
, vol. 
270
 (pg. 
9121
-
9128
)
6
Arai
H.
Monteclaro
F. S.
Tsou
C. L.
Franci
C.
Charo
I. F.
Dissociation of chemotaxis from agonist-induced receptor internalization in a lymphocyte cell line transfected with CCR2B. Evidence that directed migration does not require rapid modulation of signaling at the receptor level
J. Biol. Chem.
1997
, vol. 
272
 (pg. 
25037
-
25042
)
7
Gosling
J.
Monteclaro
F. S.
Atchison
R. E.
Arai
H.
Tsou
C. L.
Goldsmith
M. A.
Charo
I. F.
Molecular uncoupling of C-C chemokine receptor 5-induced chemotaxis and signal transduction from HIV-1 coreceptor activity
Proc. Natl. Acad. Sci. U.S.A.
1997
, vol. 
94
 (pg. 
5061
-
5066
)
8
Kraft
K.
Olbrich
H.
Majoul
I.
Mack
M.
Proudfoot
A.
Oppermann
M.
Characterization of sequence determinants within the carboxyl-terminal domain of chemokine receptor CCR5 that regulate signaling and receptor internalization
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
34408
-
34418
)
9
Venkatesan
S.
Petrovic
A.
Locati
M.
Kim
Y. O.
Weissman
D.
Murphy
P. M.
A membrane-proximal basic domain and cysteine cluster in the C-terminal tail of CCR5 constitute a bipartite motif critical for cell surface expression
J. Biol. Chem.
2001
, vol. 
276
 (pg. 
40133
-
40145
)
10
Sai
J.
Fan
G. H.
Wang
D.
Richmond
A.
The C-terminal domain LLKIL motif of CXCR2 is required for ligand-mediated polarization of early signals during chemotaxis
J. Cell Sci.
2004
, vol. 
117
 (pg. 
5489
-
5496
)
11
Terashima
Y.
Onai
N.
Murai
M.
Enomoto
M.
Poonpiriya
V.
Hamada
T.
Motomura
K.
Suwa
M.
Ezaki
T.
Haga
T.
, et al. 
Pivotal function for cytoplasmic protein FROUNT in CCR2-mediated monocyte chemotaxis
Nat. Immunol.
2005
, vol. 
6
 (pg. 
827
-
835
)
12
Toda
E.
Terashima
Y.
Sato
T.
Hirose
K.
Kanegasaki
S.
Matsushima
K.
FROUNT is a common regulator of CCR2 and CCR5 signaling to control directional migration
J. Immunol.
2009
, vol. 
183
 (pg. 
6387
-
6394
)
13
Murai
M.
Yoneyama
H.
Harada
A.
Yi
Z.
Vestergaard
C.
Guo
B.
Suzuki
K.
Asakura
H.
Matsushima
K.
Active participation of CCR5+CD8+ T lymphocytes in the pathogenesis of liver injury in graft-versus-host disease
J. Clin. Invest.
1999
, vol. 
104
 (pg. 
49
-
57
)
14
Boring
L.
Gosling
J.
Cleary
M.
Charo
I. F.
Decreased lesion formation in CCR2−/− mice reveals a role for chemokines in the initiation of atherosclerosis
Nature
1998
, vol. 
394
 (pg. 
894
-
897
)
15
Fife
B. T.
Huffnagle
G. B.
Kuziel
W. A.
Karpus
W. J.
CC chemokine receptor 2 is critical for induction of experimental autoimmune encephalomyelitis
J. Exp. Med.
2000
, vol. 
192
 (pg. 
899
-
905
)
16
Izikson
L.
Klein
R. S.
Charo
I. F.
Weiner
H. L.
Luster
A. D.
Resistance to experimental autoimmune encephalomyelitis in mice lacking the CC chemokine receptor (CCR)2
J. Exp. Med.
2000
, vol. 
192
 (pg. 
1075
-
1080
)
17
Esaki
K.
Terashima
Y.
Toda
E.
Yoshinaga
S.
Araki
N.
Matsushima
K.
Terasawa
H.
Expression and purification of human FROUNT, a common cytosolic regulator of CCR2 and CCR5
Protein Expr. Purif.
2011
, vol. 
77
 (pg. 
86
-
91
)
17a
Suwa
M.
Ono
Y.
Computational overview of GPCR gene universe to support reverse chemical genomics study
Methods Mol. Biol.
2009
, vol. 
577
 (pg. 
41
-
54
)
18
Sugihara
M.
Fujibuchi
W.
Suwa
M.
Structural elements of the signal propagation pathway in squid rhodopsin and bovine rhodopsin
J. Phys. Chem. B
2011
, vol. 
115
 (pg. 
6172
-
6179
)
19
Pontius
J.
Richelle
J.
Wodak
S. J.
Deviations from standard atomic volumes as a quality measure for protein crystal structures
J. Mol. Biol.
1996
, vol. 
264
 (pg. 
121
-
136
)
20
Zimmerman
J. M.
Eliezer
N.
Simha
R.
The characterization of amino acid sequences in proteins by statistical methods
J. Theor. Biol.
1968
, vol. 
21
 (pg. 
170
-
201
)
21
Kyte
J.
Doolittle
R. F.
A simple method for displaying the hydropathic character of a protein
J. Mol. Biol.
1982
, vol. 
157
 (pg. 
105
-
132
)
22
Fauchère
J. L.
Charton
M.
Kier
L. B.
Verloop
A.
Pliska
V.
Amino acid side chain parameters for correlation studies in biology and pharmacology
Int. J. Pept. Protein Res.
1988
, vol. 
32
 (pg. 
269
-
278
)
23
Sawano
A.
Miyawaki
A.
Directed evolution of green fluorescent protein by a new versatile PCR strategy for site-directed and semi-random mutagenesis
Nucleic Acids Res.
2000
, vol. 
28
 pg. 
E78
 
23a
Buchan
D. W. A.
Minneci
F.
Nugent
T. C. O.
Bryson
K.
Jones
D. T.
Scalable web services for the PSIPRED protein analysis workbench
Nucleic Acids Res.
2013
, vol. 
41
 (pg. 
W340
-
W348
)
24
Raman
D.
Sai
J.
Neel
N. F.
Chew
C. S.
Richmond
A.
LIM and SH3 protein-1 modulates CXCR2-mediated cell migration
PLoS ONE
2010
, vol. 
5
 pg. 
e10050
 
25
Alonzo
F.
Kozhaya
L.
Rawlings
S. A.
Reyes-Robles
T.
DuMont
A. L.
Myszka
D. G.
Landau
N. R.
Unutmaz
D.
Torres
V. J.
CCR5 is a receptor for Staphylococcus aureus leukotoxin ED
Nature
2013
, vol. 
493
 (pg. 
51
-
55
)
26
Okuno
T.
Ago
H.
Terawaki
K.
Miyano
M.
Shimizu
T.
Yokomizo
T.
Helix 8 of the leukotriene B4 receptor is required for the conformational change to the low affinity state after G-protein activation
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
41500
-
41509
)
27
Verzijl
D.
Pardo
L.
van Dijk
M.
Gruijthuijsen
Y. K.
Jongejan
A.
Timmerman
H.
Nicholas
J.
Schwarz
M.
Murphy
P. M.
Leurs
R.
Smit
M. J.
Helix 8 of the viral chemokine receptor ORF74 directs chemokine binding
J. Biol. Chem.
2006
, vol. 
281
 (pg. 
35327
-
35335
)
28
Feierler
J.
Wirth
M.
Welte
B.
Schüssler
S.
Jochum
M.
Faussner
A.
Helix 8 plays a crucial role in bradykinin B2 receptor trafficking and signaling
J. Biol. Chem.
2011
, vol. 
286
 (pg. 
43282
-
43293
)
29
Horuk
R.
Chemokine receptor antagonists: overcoming developmental hurdles
Nat. Rev. Drug Discov.
2009
, vol. 
8
 (pg. 
23
-
33
)
30
Matsushima
K.
Morishita
K.
Yoshimura
T.
Lavu
S.
Kobayashi
Y.
Lew
W.
Appella
E.
Kung
H. F.
Leonard
E. J.
Oppenheim
J. J.
Molecular cloning of a human monocyte-derived neutrophil chemotactic factor (MDNCF) and the induction of MDNCF mRNA by interleukin 1 and tumor necrosis factor
J. Exp. Med.
1988
, vol. 
167
 (pg. 
1883
-
1893
)
31
Yoshimura
T.
Yuhki
N.
Moore
S. K.
Appella
E.
Lerman
M. I.
Leonard
E. J.
Human monocyte chemoattractant protein-1 (MCP-1). Full-length cDNA cloning, expression in mitogen-stimulated blood mononuclear leukocytes, and sequence similarity to mouse competence gene JE
FEBS Lett.
1989
, vol. 
244
 (pg. 
487
-
493
)
32
Furutani
Y.
Nomura
H.
Notake
M.
Oyamada
Y.
Fukui
T.
Yamada
M.
Larsen
C. G.
Oppenheim
J. J.
Matsushima
K.
Cloning and sequencing of the cDNA for human monocyte chemotactic and activating factor (MCAF)
Biochem. Biophys. Res. Commun.
1989
, vol. 
159
 (pg. 
249
-
255
)
33
Belema-Bedada
F.
Uchida
S.
Martire
A.
Kostin
S.
Braun
T.
Efficient homing of multipotent adult mesenchymal stem cells depends on FROUNT-mediated clustering of CCR2
Cell Stem Cell.
2008
, vol. 
2
 (pg. 
566
-
575
)
34
van Golen
K. L.
Ying
C.
Sequeira
L.
Dubyk
C. W.
Reisenberger
T.
Chinnaiyan
A. M.
Pienta
K. J.
Loberg
R. D.
CCL2 induces prostate cancer transendothelial cell migration via activation of the small GTPase Rac
J. Cell Biochem.
2008
, vol. 
104
 (pg. 
1587
-
1597
)

Supplementary data